Root turnover is important to the global carbon budget as well as to nutrient cycling in ecosystems and to the success of individual plants. Our ability to predict the effects of environmental change on root turnover is limited by the difficulty of measuring root dynamics, but emerging evidence suggests that roots, like leaves, possess suites of interrelated traits that are linked to their life span. In graminoids, high tissue density has been linked to increased root longevity. Other studies have found root longevity to be positively correlated with mycorrhizal colonization and negatively correlated with nitrogen concentration, root maintenance respiration and specific root length. Among fruit trees, apple roots (which are of relatively small diameter, low tissue density and have little lignification of the exodermis) have much shorter life spans than the roots of citrus, which have opposite traits. Likewise, within the branched network of the fine root system, the finest roots with no daughter roots tend to have higher N concentrations, faster maintenance respiration, higher specific root length and shorter life spans than secondary and tertiary roots that bear daughter roots. Mycorrhizal colonization can enhance root longevity by diverse mechanisms, including enhanced tolerance of drying soil and enhanced defence against root pathogens. Many variables involved in building roots might affect root longevity, including root diameter, tissue density, N concentration, mycorrhizal fungal colonization and accumulation of secondary phenolic compounds. These root traits are highly plastic and are strongly affected by resource supply (CO2, N, P and water). Therefore the response of root longevity to altered resource availability associated with climate change can be estimated by considering how changes in resource availability affect root construction and physiology. A cost–benefit approach to predicting root longevity assumes that a plant maintains a root only until the efficiency of resource acquisition is maximized. Using an efficiency model, we show that reduced tissue Nconcentration and reduced root maintenance respiration, both of which are predicted to result from elevated CO2, should lead to slightly longer root life spans. Complex interactions with soil biota and shifts in plant defences against root herbivory and parasitism, which are not included in the present efficiency model, might alter the effects of future climate change on root longevity in unpredicted ways.